Incremental Top-k List Comparison Approach to Robust Multi-Structure Model Fitting

TL;DR

This paper introduces a novel hypothesis sampling method based on incremental top-k list comparisons, enhancing the efficiency and accuracy of multi-structure model fitting in robust estimation tasks.

Contribution

It presents a new incremental ranking-based sampling scheme that improves hypothesis retrieval and filtering for multi-structure model fitting.

Findings

Outperforms previous methods on synthetic data

Demonstrates superior accuracy on real datasets

Achieves higher efficiency in hypothesis filtering

Abstract

Random hypothesis sampling lies at the core of many popular robust fitting techniques such as RANSAC. In this paper, we propose a novel hypothesis sampling scheme based on incremental computation of distances between partial rankings (top-$k$ lists) derived from residual sorting information. Our method simultaneously (1) guides the sampling such that hypotheses corresponding to all true structures can be quickly retrieved and (2) filters the hypotheses such that only a small but very promising subset remain. This permits the usage of simple agglomerative clustering on the surviving hypotheses for accurate model selection. The outcome is a highly efficient multi-structure robust estimation technique. Experiments on synthetic and real data show the superior performance of our approach over previous methods.